check-hardcode

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious behavior or security vulnerabilities detected. The skill is purely instructional and serves as a best-practice guide for code reviews.
  • [PROMPT_INJECTION]: The skill processes untrusted code, which creates a surface for indirect prompt injection. However, the risk is minimal as the skill lacks access to sensitive tools or network operations.
  • Ingestion points: Code provided by the user for hardcoding checks in references/checklist.md
  • Boundary markers: None
  • Capability inventory: No external tools or system access capabilities
  • Sanitization: None
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 11:57 AM
Security Audit — agent-trust-hub — check-hardcode